99.98% chance that Sen. Barack Obama will win. (Franklin & Marshall study)

[BurgherHoldtheLies]

http://politicalwire.com/archives/2008/10/31/simulation...

Simulation Shows Obama Will Win

Brian Adams, a mathematics and computer science professor at Franklin & Marshall College, reports that there's a 99.98% chance that Sen. Barack Obama will win the presidential election on Tuesday.

Adams has developed a simulation model that involves running 50 million simulated state-by-state races, using the late-October poll results for each state. He ran his simulation 50 million times to allow him to calculate all the different combinations of electoral votes that might result, even long shots. While the number of possible outcomes is very large, the result is always the same. Obama receives 270 or more electoral votes 99.98% of the time. Using intervals of electoral vote results, there is a 95 percent chance that the outcome will have Obama winning between 303 and 381 electoral votes.

October 31, 2008

[Gman]

1. Brian Adams' model correctly states that Obama wins in late October

but the election is not until November. He's using data that is a "duh..." thing. I think we all know what the outcome would be if the election were this week. But the polls seem to be moving right now.

[FrenchieCat]

2. Are they really? Or are folks just saying that they are over and over again?

and using different polls to show movement.....like PA poll from one company one day, and then using a different polling company the next and claiming that there is a...how do they say it? t.i.g.h.t.e.n.i.n.g.

Love the way they say it....

[Schulzz]

3. In my opinion, McCain would need the biggest fraud in history to win.

He would have to steal several states where Obama is leading outside the MoE. Florida or Ohio just won't do it.

[Ioo]

5. And the Republicans do not like him so much as to help him pull it off

They would try for her, not him.

[sarahdemva]

4. what is his track record for past elections?

[Faygo Kid]

6. I just KNEW that .02% would be there. Another reason for concern.

[Jackpine Radical]

7. Is he polling the ES&S machines? How's the Debold vote doing?

How about those challenge lists?

No model is any better than the data you put into it.